Publication Details

Category Text Publication
Reference Category Journals
DOI 10.1017/eds.2022.1
Licence creative commons licence
Title (Primary) Predicting years with extremely low gross primary production from daily weather data using Convolutional Neural Networks
Author Marcolongo, A.; Vladymyrov, M.; Lienert, S.; Peleg, N.; Haug, S.; Zscheischler, J. ORCID logo
Source Titel Environmental Data Science
Year 2022
Department CHS
Volume 1
Page From e2
Language englisch
Topic T5 Future Landscapes
Data and Software links http://doi.org/10.5281/zenodo.5555266
Supplements http://doi.org/10.1017/eds.2022.1
Keywords Carbon cycle; Convolutional Neural Networks; extreme events; extreme impacts
Abstract

Understanding the meteorological drivers of extreme impacts in social or environmental systems is important to better quantify current and project future climate risks. Impacts are typically an aggregated response to many different interacting drivers at various temporal scales, rendering such driver identification a challenging task. Machine learning–based approaches, such as deep neural networks, may be able to address this task but require large training datasets. Here, we explore the ability of Convolutional Neural Networks (CNNs) to predict years with extremely low gross primary production (GPP) from daily weather data in three different vegetation types. To circumvent data limitations in observations, we simulate 100,000 years of daily weather with a weather generator for three different geographical sites and subsequently simulate vegetation dynamics with a complex vegetation model. For each resulting vegetation distribution, we then train two different CNNs to classify daily weather data (temperature, precipitation, and radiation) into years with extremely low GPP and normal years. Overall, prediction accuracy is very good if the monthly or yearly GPP values are used as an intermediate training target (area under the precision-recall curve AUC https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220412092720997-0993:S2634460222000012:S2634460222000012_inline1.png0.9). The best prediction accuracy is found in tropical forests, with temperate grasslands and boreal forests leading to comparable results. Prediction accuracy is strongly reduced when binary classification is used directly. Furthermore, using daily GPP during training does not improve the predictive power. We conclude that CNNs are able to predict extreme impacts from complex meteorological drivers if sufficient data are available.

Persistent UFZ Identifier https://www.ufz.de/index.php?en=20939&ufzPublicationIdentifier=26224
Marcolongo, A., Vladymyrov, M., Lienert, S., Peleg, N., Haug, S., Zscheischler, J. (2022):
Predicting years with extremely low gross primary production from daily weather data using Convolutional Neural Networks
Environ. Data Sci. 1 , e2 10.1017/eds.2022.1